arXiv subjects

Gyemin Lee

Publications and source records attributed to Gyemin Lee.

4 recordsLinked to original sources

arXiv2023

Feature Alignment by Uncertainty and Self-Training for Source-Free Unsupervised Domain Adaptation

Most unsupervised domain adaptation (UDA) methods assume that labeled source images are available during model adaptation. However, this assumption is often infeasible owing to confidentiality issues or memory constraints on mobile devices. Some recently developed approaches do not require source images during adaptation, but they show limited performance on perturbed images. To address these problems, we propose a novel source-free UDA method that uses only a pre-trained source model and unlabeled target images. Our method captures the aleatoric uncertainty by incorporating data augmentation and trains the feature generator with two consistency objectives. The feature generator is encouraged to learn consistent visual features away from the decision boundaries of the head classifier. Thus, the adapted model becomes more robust to image perturbations. Inspired by self-supervised learning, our method promotes inter-space alignment between the prediction space and the feature space while incorporating intra-space consistency within the feature space to reduce the domain gap between the source and target domains. We also consider epistemic uncertainty to boost the model adaptation performance. Extensive experiments on popular UDA benchmark datasets demonstrate that the proposed source-free method is comparable or even superior to vanilla UDA methods. Moreover, the adapted models show more robust results when input images are perturbed.

cs.CV↗

arXiv2022

Unsupervised Domain Adaptation Based on the Predictive Uncertainty of Models

Unsupervised domain adaptation (UDA) aims to improve the prediction performance in the target domain under distribution shifts from the source domain. The key principle of UDA is to minimize the divergence between the source and the target domains. To follow this principle, many methods employ a domain discriminator to match the feature distributions. Some recent methods evaluate the discrepancy between two predictions on target samples to detect those that deviate from the source distribution. However, their performance is limited because they either match the marginal distributions or measure the divergence conservatively. In this paper, we present a novel UDA method that learns domain-invariant features that minimize the domain divergence. We propose model uncertainty as a measure of the domain divergence. Our UDA method based on model uncertainty (MUDA) adopts a Bayesian framework and provides an efficient way to evaluate model uncertainty by means of Monte Carlo dropout sampling. Empirical results on image recognition tasks show that our method is superior to existing state-of-the-art methods. We also extend MUDA to multi-source domain adaptation problems.

cs.CV↗

arXiv2021

Domain Generalization by Marginal Transfer Learning

In the problem of domain generalization (DG), there are labeled training data sets from several related prediction problems, and the goal is to make accurate predictions on future unlabeled data sets that are not known to the learner. This problem arises in several applications where data distributions fluctuate because of environmental, technical, or other sources of variation. We introduce a formal framework for DG, and argue that it can be viewed as a kind of supervised learning problem by augmenting the original feature space with the marginal distribution of feature vectors. While our framework has several connections to conventional analysis of supervised learning algorithms, several unique aspects of DG require new methods of analysis. This work lays the learning theoretic foundations of domain generalization, building on our earlier conference paper where the problem of DG was introduced (Blanchard et al., 2011). We present two formal models of data generation, corresponding notions of risk, and distribution-free generalization error analysis. By focusing our attention on kernel methods, we also provide more quantitative results and a universally consistent algorithm. An efficient implementation is provided for this algorithm, which is experimentally compared to a pooling strategy on one synthetic and three real-world data sets.

stat.ML↗

arXiv2010

Statistical File Matching of Flow Cytometry Data

Flow cytometry is a technology that rapidly measures antigen-based markers associated to cells in a cell population. Although analysis of flow cytometry data has traditionally considered one or two markers at a time, there has been increasing interest in multidimensional analysis. However, flow cytometers are limited in the number of markers they can jointly observe, which is typically a fraction of the number of markers of interest. For this reason, practitioners often perform multiple assays based on different, overlapping combinations of markers. In this paper, we address the challenge of imputing the high dimensional jointly distributed values of marker attributes based on overlapping marginal observations. We show that simple nearest neighbor based imputation can lead to spurious subpopulations in the imputed data, and introduce an alternative approach based on nearest neighbor imputation restricted to a cell's subpopulation. This requires us to perform clustering with missing data, which we address with a mixture model approach and novel EM algorithm. Since mixture model fitting may be ill-posed, we also develop techniques to initialize the EM algorithm using domain knowledge. We demonstrate our approach on real flow cytometry data.

stat.AP↗